Irrelevant Objects Research Articles

Autistic traits and perspective taking in youths with anorexia nervosa: an exploratory clinical and eye tracking study

BackgroundDespite their apparent dissimilarity, Anorexia Nervosa (AN) and Autism Spectrum Disorder (ASD) share many features, especially in terms of social and emotional difficulties. In recent years, empathic abilities in AN have been frequently assessed using self-report measures. Otherwise, the director task (DT) has been used to investigate the ability to take the visual perspective of another individual in a communicative context, using eye-tracking technology. The aim of the current study was to test the presence of autism-relevant features in AN, through: (i) comparing self-reported autistic traits and empathic abilities in a group of young inpatients with AN and age/gender matched healthy controls (HC); (ii) comparing performance on the director paradigm.MethodsThe participants were females in the age-range between 11 and 18 years: 24 with AN and 23 HC. Autistic traits, empathic abilities, and severity of the eating disorder were respectively measured using: the Autism Quotient (AQ), the Interpersonal Reactivity Index (IRI), and the Eating Disorder Inventory-3 (EDI-3). Both groups performed a computerized task in which a director instructed them to move objects placed on a set of shelves using a mouse, while their eye gaze was tracked. A total of 36 shelf configurations, divided into three categories (with dimensional distractor – with spatial distractor – control), were created.ResultsSubjects with AN showed higher autistic traits than HC. Eye-tracking data revealed that subjects with AN took longer to decide which object to select and where to move it, both in distractor-trials and in control-trials. In the AN group, we found a significant negative correlation between the total score of the AQ and the number of fixations to the irrelevant object in the dimensional control condition -in which the subjects were asked to focus on dimensional aspects of the object (large-small)-.ConclusionsAutistic traits were over-represented in a group of young inpatients with AN. Through the use of eye-tracking technology, this exploratory study documented some differences between AN inpatients and HC in their online processes during the perspective taking tasks, which could be considered a target of tailored intervention. A larger sample of patients is needed to confirm these preliminary findings.

TopicRefiner: Coherence-Guided Steerable LDA for Visual Topic Enhancement.

This article presents a new Human-steerable Topic Modeling (HSTM) technique. Unlike existing techniques commonly relying on matrix decomposition-based topic models, we extend LDA as the fundamental component for extracting topics. LDA's high popularity and technical characteristics, such as better topic quality and no need to cherry-pick terms to construct the document-term matrix, ensure better applicability. Our research revolves around two inherent limitations of LDA. First, the principle of LDA is complex. Its calculation process is stochastic and difficult to control. We thus give a weighting method to incorporate users' refinements into the Gibbs sampling to control LDA. Second, LDA often runs on a corpus with massive terms and documents, forming a vast search space for users to find semantically relevant or irrelevant objects. We thus design a visual editing framework based on the coherence metric, proven to be the most consistent with human perception in assessing topic quality, to guide users' interactive refinements. Cases on two open real-world datasets, participants' performance in a user study, and quantitative experiment results demonstrate the usability and effectiveness of the proposed technique.

Semantic 3D reconstruction-oriented image dataset for building component segmentation

In image-driven 3D building reconstruction, instance segmentation is fundamental to pixel-wise building component detection, which can be fused with 3D data like point clouds and meshes via camera projection for semantic reconstruction. While deep learning-based segmentation has obtained promising results, it relies heavily on large-scale datasets for training. Unfortunately, existing large-scale image datasets often include irrelevant objects that obstruct building components, making them unsuitable for 3D building reconstruction. This paper addresses this gap by introducing a large-scale building image dataset to facilitate building component segmentation for 3D reconstruction. The dataset comprises 3378 images captured from both interiors and exteriors of 36 university buildings, annotated with 49,380 object instances across 11 classes. Rigorous quality control measures were employed during data collection and annotation. Evaluation of five typical deep learning-based instance segmentation models demonstrates the dataset's suitability for training and its value as a benchmark dataset for building component segmentation.

Machine Learning for the Visually Impaired: Benchmarking Object Detection Models

This research paper benchmarks object detection models, a form of machine learning, to determine which algorithm would be most beneficial for the visually impaired to locate objects that are used in everyday life. As most benchmarking experiments test object detection models with a variety of objects, it is essential to test the models using images of more relevant objects to find the most suitable algorithm. The models are tested using still images from the COCO database. Pretrained models employing five of the most popular object detection algorithms are used to process the images and find each model’s detection accuracy. To simulate real life scenarios, these objects may be partially hidden or at a distance. For each image, the models return a list of detections providing the names, confidence rating, and location of each object detected. These results will be filtered to remove detections with low confidence ratings as well as detections of irrelevant objects. The remaining results are compared to the dataset of object names and locations provided by the COCO database to calculate the distance between the predicted object locations and the true location. The algorithms will be ranked based on the number of failed detections, the time taken to analyze each image, and the accuracy of each object’s determined location.

Detection-Based Intermediate Supervision for Visual Question Answering

Recently, neural module networks (NMNs) have yielded ongoing success in answering compositional visual questions, especially those involving multi-hop visual and logical reasoning. NMNs decompose the complex question into several sub-tasks using instance-modules from the reasoning paths of that question and then exploit intermediate supervisions to guide answer prediction, thereby improving inference interpretability. However, their performance may be hindered due to sketchy modeling of intermediate supervisions. For instance, (1) a prior assumption that each instance-module refers to only one grounded object yet overlooks other potentially associated grounded objects, impeding full cross-modal alignment learning; (2) IoU-based intermediate supervisions may introduce noise signals as the bounding box overlap issue might guide the model's focus towards irrelevant objects. To address these issues, a novel method, Detection-based Intermediate Supervision (DIS), is proposed, which adopts a generative detection framework to facilitate multiple grounding supervisions via sequence generation. As such, DIS offers more comprehensive and accurate intermediate supervisions, thereby boosting answer prediction performance. Furthermore, by considering intermediate results, DIS enhances the consistency in answering compositional questions and their sub-questions. Extensive experiments demonstrate the superiority of our proposed DIS, showcasing both improved accuracy and state-of-the-art reasoning consistency compared to prior approaches.

AMFF-Net: An attention-based multi-scale feature fusion network for allergic pollen detection

Automatic pollen detection based on light microscope (LM) images is helpful for pollinosis symptoms prevention. Recently, many deep learning methods have been proposed to identify pollen grains based on multi-scale feature fusion mechanism. However, in real scenarios, there are two main challenges that need to be considered: (1) Complex pollen characteristics; (2) Irrelevant objects interference. It means that the pollen detection requires not only learning the relationship among multi-scale features but also refining the feature representation. To this end, this paper proposes an attention-based multi-scale feature fusion network (AMFF-Net) for automatic pollen detection on real-world LM images. The proposed AMFF-Net includes three modules: The feature extraction module utilizes the series-connection attention to capture the spatial and channel dependencies of different level feature maps (for solving the challenge 1). In the feature fusion module, a parallel connection attention is able to learn more discriminative feature representation based on bidirectional pathway guidance (for solving the challenge 2). Both are jointly adopted to enhance the representational capacity of the final results in the pollen prediction module. Extensive experiments are conducted on the real-world RPD dataset, and our AMFF-Net achieves the best performance (83.9% of mean average precision) comparing with other state-of-the-art methods. We believe that this work can serve as an important reference for the development of pollen monitoring system in a real scenario.

Distractor's salience does not determine feature suppression: A commentary on Wang and Theeuwes (2020).

There is ongoing debate as to whether distraction by salient irrelevant objects can be avoided by suppressing their salient features. Lamy suggested that a central reason for this stalemate is methodological: researchers often base their conclusions on whether the presence of the salient distractor yields net interference (interpreted as capture) or benefit (interpreted as suppression), instead of investigating how manipulating inhibitory suppression modulates these effects. Here, we validate this observation by revisiting Wang and Theeuwes' findings showing that a color singleton distractor produced a net cost with dense search displays and a net benefit with sparse displays. They concluded that only mildly salient distractors can be suppressed. In two experiments, we orthogonally manipulated distractor salience and feature-based suppression. Participants searched for a shape and a color singleton was sometimes present. Search displays were either sparse or dense and the singleton's color changed on each block. Distractor feature-based suppression was measured as a reduction in distractor interference in the second relative to the first half of each block. We replicated Wang and Theeuwes' findings but invalidated their interpretation by showing that participants learned to suppress the color singleton equally well when displays were sparse and when they were dense. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Semi-Supervised Semantic Segmentation for Identification of Irrelevant Objects in a Waste Recycling Plant

In waste recycling plants, measuring the waste volume and weight at the beginning of the treatment process is key for a better management of resources. This task can be conducted by using orthophoto images, but it is necessary to remove from those images the objects, such as containers or trucks, that are not involved in the measurement process. This work proposes the application of deep learning for the semantic segmentation of those irrelevant objects. Several deep architectures are trained and compared, while three semi-supervised learning methods (PseudoLabeling, Distillation and Model Distillation) are proposed to take advantage of non-annotated images. In these experiments, the U-net++ architecture with an EfficientNetB3 backbone, trained with the set of labelled images, achieves the best overall multi Dice score of 91.23%. The application of semi-supervised learning methods further boosts the segmentation accuracy in a range between 1.31% and 2.59%, on average.

Editorial

Dear Readers,  I am very happy to announce the fifth regular issue of 2023. In this issue, 5 articles by 22 authors from 5 countries cover a variety of topical research aspects in computer science. Allow me to express my appreciation to all authors for their sound research and to the editorial board for the highly valuable reviews and comments for improvement. This continuous stream of relevant and novel contributions, along with the generous support of the consortium members, sustains the quality of our journal.  In the ongoing effort to further strengthen our journal, I would like to expand the editorial board: If you are a tenured associate professor or above with a strong publication record, you are welcome to apply to join our editorial board. We are also interested in high-quality proposals for special issues on new topics and trends. Please consider yourself and encourage your colleagues to submit high-quality articles or special issues for our journal.  In the fifth regular issue, I am very pleased to introduce the following five accepted articles: César Domínguez Pérez, Jónathan Heras, Eloy Mata, Vico Pascual, Lucas Fernández-Cedrón, Marcos Martínez-Lanchares, Jon Pellejero-Espinosa, Antonio Rubio-Loscertales, and Carlos Tarragona-Perez from Spain report on their deep learning approach for semi-supervised semantic segmentation to identify irrelevant objects in a waste recycling plant. In a research collaboration between Australia and Iraq, Mitchell Jensen, Khamael Al-Dulaimi, Khairiyah Saeed Abduljabbar and Jasmine Banks are focusing on their work on autoimmune disease detection in humans, more specifically on automating the classification procedure of HEp-2 stained cells from microscopic images and improving the accuracy of computer-aided diagnosis. Kashif Mehboob Khan, Warda Haider, Najeed Ahmed Khan, and Darakhshan Saleem from Pakistan address their research on big data provenance using blockchain for qualitative analytics through machine learning. Abubakhari Sserwadda, Alper Ozcan, and Yusuf Yaslan from Türkiye present their research on a novel end-to-end unified topological similarity and centrality driven hybrid deep learning model for temporal link prediction. And last but not least, Tahseen A. Wotaifi, and Ban N. Dhannoon from Iraq aim in their research to use deep learning, pre-trained models, and machine learning based on Convolution Neural Networks to predict Arabic and English fake news based on three public and available datasets: the Fake-or-Real dataset, the AraNews dataset, and the Sentimental LIAR dataset. Enjoy Reading! Cordially,  Christian Gütl, Managing Editor

M1-selective muscarinic allosteric modulation enhances cognitive flexibility and effective salience in nonhuman primates

Acetylcholine (ACh) in cortical neural circuits mediates how selective attention is sustained in the presence of distractors and how flexible cognition adjusts to changing task demands. The cognitive domains of attention and cognitive flexibility might be differentially supported by the M1 muscarinic acetylcholine receptor (mAChR) subtype. Understanding how M1 mAChR mechanisms support these cognitive subdomains is of highest importance for advancing novel drug treatments for conditions with altered attention and reduced cognitive control including Alzheimer's disease or schizophrenia. Here, we tested this question by assessing how the subtype-selective M1 mAChR positive allosteric modulator (PAM) VU0453595 affects visual search and flexible reward learning in nonhuman primates. We found that allosteric potentiation of M1 mAChRs enhanced flexible learning performance by improving extradimensional set shifting, reducing latent inhibition from previously experienced distractors and reducing response perseveration in the absence of adverse side effects. These procognitive effects occurred in the absence of apparent changes of attentional performance during visual search. In contrast, nonselective ACh modulation using the acetylcholinesterase inhibitor (AChEI) donepezil improved attention during visual search at doses that did not alter cognitive flexibility and that already triggered gastrointestinal cholinergic side effects. These findings illustrate that M1 mAChRpositive allosteric modulation enhances cognitive flexibility without affecting attentional filtering of distraction, consistent with M1 activity boosting the effective salience of relevant over irrelevant objects specifically during learning. These results suggest that M1 PAMs are versatile compounds for enhancing cognitive flexibility in disorders spanning schizophrenia and Alzheimer's diseases.

Recovering Microscopic Images in Material Science Documents by Image Inpainting

Microscopic images in material science documents have increased in number due to the growth and common use of electron microscopy instruments. Through the use of data mining techniques, they are easily accessible and can be obtained from documents published online. As data-driven approaches are becoming increasingly common in the material science field, massively acquired experimental images through microscopy play important roles in terms of developing an artificial intelligence (AI) model for the purposes of automatically diagnosing crucial material structures. However, irrelevant objects (e.g., letters, scale bars, and arrows) that are often present inside original microscopic photos should be removed for the purposes of improving the AI models. To avoid the issue above, we applied four image inpainting algorithms (i.e., shift-net, global and local, contextual attention, and gated convolution) to a learning approach, with the aim of recovering microscopic images in journal papers. We estimated the structural similarity index measure (SSIM) and ℓ1/ℓ2 errors, which are often used as measures of image quality. Lastly, we observed that gated convolution possessed the best performance for inpainting the microscopic images.

THAT-Net: Two-layer hidden state aggregation based two-stream network for traffic accident prediction

Traffic accident prediction based on dashboard cameras is of great significance for ensuring the guaranteed-safety of self-driving systems and reducing the occurrence of accidents. Due to complexity of traffic scenarios and wide range of object's motion, early prediction of accidents remains a challenge. However, the existing methods process all traffic objects and all video frames indiscriminately, thus leading to poor prediction performance. To address this issue, we propose a novel traffic accident prediction framework, namely two-layer hidden state aggregation based two-stream network (THAT-Net). The proposed method first fuses the spatial and temporal flow to capture complementary spatio-temporal information, which filters out irrelevant objects in traffic scenes to reduce the influence of the object's motion state. Furthermore, a two-layer hidden state aggregation structure is designed to reintegrate the hidden state weights of gated recurrent units. It captures contextual information through frame-level and segment-level aggregation and decreases the influence of irrelevant frames to reduce the complexity of traffic scenes. Experiments based on two real-world datasets show the state-of-the-art performance of THAT-Net. Our proposed method achieves the highest accuracy by predicting accidents 0.48 sec to 2.8 sec earlier compared to the baseline methods in more challenging situations. Our code is available at: https://github.com/redeyezt/THAT-Net.

Interaction of dynamic error signals in saccade adaptation.

Motor adaptation maintains movement accuracy. To evaluate movement accuracy, motor adaptation relies on an error signal, generated by the movement target, while suppressing error signals from irrelevant objects in the vicinity. Previous work used static testing environments, where all information required to evaluate movement accuracy was available simultaneously. Using saccadic eye movements as a model for motor adaptation, we tested how movement accuracy is maintained in dynamic environments, where the availability of conflicting error signals varied over time. Participants made a vertical saccade toward a target (either a small square or a large ring). Upon saccade detection, two candidate stimuli were shown left and right of the target, and participants were instructed to discriminate a feature on one of the candidates. Critically, candidate stimuli were presented sequentially, and saccade adaptation, thus, had to resolve a conflict between a task-relevant and a task-irrelevant error signal that were separated in space and time. We found that the saccade target influenced several aspects of oculomotor learning. In presence of a small target, saccade adaptation evaluated movement accuracy based on the first available error signal after the saccade, irrespective of its task relevance. However, a large target not only allowed for greater flexibility when evaluating movement accuracy, but it also promoted a stronger contribution of strategic behavior when compensating inaccurate saccades. Our results demonstrate how motor adaptation maintains movement accuracy in dynamic environments, and how properties of the visual environment modulate the relative contribution of different learning processes.NEW & NOTEWORTHY Motor adaptation is typically studied in static environments, where all information that is required to evaluate movement accuracy is available simultaneously. Here, using saccadic eye movements as a model, we studied motor adaptation in a dynamic environment, where the availability of conflicting information about movement accuracy varied over time. We demonstrate that properties of the visual environment determine how dynamic movement errors are corrected.

Referring Expression Comprehension Via Enhanced Cross-modal Graph Attention Networks

Referring expression comprehension aims to localize a specific object in an image according to a given language description. It is still challenging to comprehend and mitigate the gap between various types of information in the visual and textual domains. Generally, it needs to extract the salient features from a given expression and match the features of expression to an image. One challenge in referring expression comprehension is the number of region proposals generated by object detection methods is far more than the number of entities in the corresponding language description. Remarkably, the candidate regions without described by the expression will bring a severe impact on referring expression comprehension. To tackle this problem, we first propose a novel Enhanced Cross-modal Graph Attention Networks (ECMGANs) that boosts the matching between the expression and the entity position of an image. Then, an effective strategy named Graph Node Erase (GNE) is proposed to assist ECMGANs in eliminating the effect of irrelevant objects on the target object. Experiments on three public referring expression comprehension datasets show unambiguously that our ECMGANs framework achieves better performance than other state-of-the-art methods. Moreover, GNE is able to obtain higher accuracies of visual-expression matching effectively.

A Novel Deep Nearest Neighbor Neural Network for Few-Shot Remote Sensing Image Scene Classification

Remote sensing image scene classification has become more and more popular in recent years. As we all know, it is very difficult and time-consuming to obtain a large number of manually labeled remote sensing images. Therefore, few-shot scene classification of remote sensing images has become an urgent and important research task. Fortunately, the recently proposed deep nearest neighbor neural network (DN4) has made a breakthrough in few-shot classification. However, due to the complex background in remote sensing images, DN4 is easily affected by irrelevant local features, so DN4 cannot be directly applied in remote sensing images. For this reason, a deep nearest neighbor neural network based on attention mechanism (DN4AM) is proposed to solve the few-shot scene classification task of remote sensing images in this paper. Scene class-related attention maps are used in our method to reduce interference from scene-semantic irrelevant objects to improve the classification accuracy. Three remote sensing image datasets are used to verify the performance of our method. Compared with several state-of-the-art methods, including MatchingNet, RelationNet, MAML, Meta-SGD and DN4, our method achieves promising results in the few-shot scene classification of remote sensing images.

The effect of memory load on object reconstruction: Insights from an online mouse-tracking task.

Why can't we remember everything that we experience? Previous work in the domain of object memory has suggested that our ability to resolve interference between relevant and irrelevant object features may limit how much we can remember at any given moment. Here, we developed an online mouse-tracking task to study how memory load influences object reconstruction, testing participants synchronously over virtual conference calls. We first tested up to 18 participants concurrently, replicating memory findings from a condition where participants were tested individually. Next, we examined how memory load influenced mouse trajectories as participants reconstructed target objects. We found interference between the contents of working memory and what was perceived during object reconstruction, an effect that interacted with visual similarity and memory load. Furthermore, we found interference from previously studied but currently irrelevant objects, providing evidence of object-to-location binding errors. At the greatest memory load, participants were nearly three times more likely to move their mouse cursor over previously studied nontarget objects, an effect observed primarily during object reconstruction rather than in the period before the final response. As evidence of the dynamic interplay between working memory and perception, these results show that object reconstruction behavior may be altered by (i) interference between what is represented in mind and what is currently being viewed, and (ii) interference from previously studied but currently irrelevant information. Finally, we discuss how mouse tracking can provide a rich characterization of participant behavior at millisecond temporal resolution, enormously increasing power in cognitive psychology experiments.

New methods for the drawing of the Archaeological Forma Urbis: Bohob’s research in Catania

Nowadays, in the centre of Catania, the several visible archaeologies seem more irrelevant objects than ancient treasures. This is due to fact they are fenced in scattered holes without any common idea. “Bohob”, the group of architects we belong to, is conducting a research in order to address this burning problem by means of a two-pronged approach. On one hand, by the Drawing of the City, a large plan of Catania including all the archaeological ruins representing their integral role in the building of the city and in forcing the shape of the modern town; on the other hand, by the Minimum Projects, consisting of shrines on a citywide scale in which the ruins are enclosed as well as appearing as gems and not as temporary objects in the urban pattern. The core essence of the research is studying the relationship between the archae-ology and the city via an interdisciplinary method based on the brand-new digital forms of representation; starting from considering the ancient as “active part” in the construction of the town and its little fragments as gems mounted in the urban pattern.

Investigating the mechanisms by which selective attention affects subsequent preferences and choice

In two experiments, we investigated two untested assumptions regarding the mechanism by which selective attention during search affects subsequent preferences for objects. First, we tested whether an increase in visual competition during search increases preferences for relevant objects and decreases preferences for irrelevant objects subsequent to search. Second, we tested whether searching for objects increases the perceived fluency to process relevant objects and decreases the perceived fluency to process irrelevant objects. Our results show that search can affect relevant and irrelevant objects differently. Selective attention increased preferences for target objects subsequent to search, whereas selective attention did not affect preferences for distractors. Furthermore, our results indicate that searching for a target object increased the perceived fluency for this target object during choice, whereas ignoring a distractor product blocked mere exposure effects. Contrary to assumptions made in previous research, we found no indication that the competition for visual resources during search is linked to preferences for targets or distractors.

DMA-Net: DeepLab With Multi-Scale Attention for Pavement Crack Segmentation

Cracks are important indicators of pavement structural and operational conditions. Early pavement crack detection and treatments can help extend pavement service life, reduce fuel consumption, and improve safety and ride quality. Pavement distress surveys have traditionally been performed manually by visually inspecting the roads, which is labor-intensive and time-consuming. Therefore, computer-vision-based automated crack detection has great practical significance in pavement maintenance and traffic safety. Traditional image processing techniques are sensitive to noise in images and are thus likely to miss detecting some cracks due to the crack texture variety, complex lighting conditions, and various similar but irrelevant objects on the road. This paper adopts and enhances DeepLabv3+, a popular deep learning framework for semantic image segmentation, for road pavement crack detection. We propose a multi-scale attention module in the decoder of DeepLabv3+ to generate an attention mask and dynamically assign weights between high-level and low-level feature maps. Compared with fixed weights across different features, the dynamic weights strategy can assign more reasonable weights to different feature maps. Ablation experiments show that the attention mask can effectively help the model better combine multi-scale features and generate more accurate pavement crack segmentation results. The proposed method achieves state-of-the-art results on three benchmarks, including Crack500, DeepCrack, and FMA (Fitchburg Municipal Airport) datasets. We further test it on pavement crack images captured by smartphones, and the results show that it provides a viable approach to road pavement crack segmentation in practice with excellent performance.

Attenuating the ‘attentional white bear’ effect enhances suppressive attention

Trying to ignore an object can bias attention towards it – a phenomenon referred to as the ‘attentional white bear’ (AWB) effect. The mechanisms behind this effect remain unclear. On one hand, the AWB may reflect reactive, ‘search and destroy’ distractor suppression, which directs attention toward irrelevant objects in order to suppress further attention to them. However, another possibility is that the AWB results from failed proactive distractor suppression – attempting to suppress attention to an irrelevant object from the outset may inadvertently result in an attentional shift towards it. To distinguish these two possibilities, we developed a categorical visual search task that addresses limitations present in prior studies. In five experiments (Ntotal = 96), participants searched displays of naturalistic stimuli cued only with distractor categories (targets were unknown and unpredictable). We observed an AWB and later attenuated it by presenting a pre-search stimulus, likely disrupting guidance from distractor templates in working memory. We conclude that the AWB resulted from a failure of proactive suppression rather than a search and destroy process.